Heterogeneous gender effects under loss aversion in the economics classroom: A field experiment

Abstract

This article evaluates the impact of loss aversion as a behavioral motivator on students classroom performance. Conducting an experiment with University of Kentucky undergraduate students, the authors framed student grades as a loss and gain. In treatment sections, the students began with full marks and lost points as the semester progressed, whereas in control sections, under a traditional grading scheme, students accumulated points throughout the semester. We find that treated individuals, on average, do not have a statistically different final grade than individuals in the control class. However, we uncover a heterogeneous gender effect. On average, a male in the treatment class scores between 3.17 and 4.05 percentage points higher on the final grade than a male in the control class, ceteris paribus. Conversely, a female in the treatment class scores between 3.61 and 4.36 percentage points lower on the final grade than a comparable female in the control class. JEL Classification: A

Nudging in education

“…McEvoy (2016) on the other hand reported no gender effects in a similar study also testing a loss framing for grade points earned by US university students. In contrast to Apostolova-Mihaylova et al (2015) and McEvoy (2016) found positive effects on grades and the likelihood of attempting the final exam when controlling for individual specific characteristics. The effect of the loss frame appeared most important for assignments at the end of the semester.…”

Improving student performance through loss aversion.

“…In the course of this investigation we also determined whether our data set showed evidence of a heterogeneous-sex effect from the loss aversion treatment when controlling for student ability, as was found in Apostolova-Mihaylova et al (2015). This can be stated as our second hypothesis:…”

“…The differential impact of the Treatment ؋ Sex specification. The work of Apostolova-Mihaylova et al (2015) indicated that female students might perform worse under the inverted grading method, compared to the male students. Using our data set, we were interested in replicating this result.…”

“…However, the study suggested that female students were negatively impacted by the treatment, whereas male students were positively impacted by the treatment. Apostolova-Mihaylova et al (2015) suffered from a number of empirical shortcomings that this study attempts to rectify. The previous studies had no mechanism to control for student domain knowledge and relied on unusual controls for mathematics ability; no attempt was made to control for selection due to the consent process.…”

Loss aversion as a motivator for engagement with peer assessment

“…Levitt et al (2016) also find some evidence that students respond more to rewards framed as losses. Of greatest relevance to the design of our assessment, Apostolova-Mihaylova et al (2015) find that framing the grading system in terms of losses leads to better outcomes, though only for males. Women in the study performed worse under the loss aversion frame.…”